Descriptive Analysis & Preliminary Results
Introduction
The following document summarises the progress made thus far on Chapter 1: Local Fiscal Risks of Decarbonisation of my DPhil. The work aims to pursue a better understanding of how industrial transformation impacts local well-being. From an original interest in looking at all aspects of local public finance, the project has narrowed to focus on expenditure on public education and its connection to industrial prosperity and transformation.
Below, I provide some descriptive statistics and figures about the data, some preliminary regression tests of the significance of various GDP control variables, and some regression results using a standard TWFE model (with and without state-level trends) and coal mining activity as instruments for property taxes and GDP.
The main research question is: How does industrial transformation/activity impact county-level expenditure on public education?
Conclusion: I do not yet feel confident in any of the results below but hopefully these steps can help orient the next phase of analysis (ideally moving on to consider oil and gas industry treatment variables as well or treatments that proxy “industrial transformation” in a more general sense).
Data
All data used is reported annually at the county-level. Therefore, no time-invariant variables are included (apart from the State in which a county is in, which is made time-variant through the inclusion of a state-level trend in various models).
Expenditure and Revenue: The dependent variables of interest come from Willamette University’s Government Finance Database. The data includes county-level revenue and expenditure on public education including disaggregated values by revenue source (federal, state, or other intergovernmental revenue) and expenditure item (lunches, wages, debt). All values are reported in current US dollars. The data for property taxes collected used in regressions below also come from this dataset. Expenditure on vocational training and from Educational Service Agencies (ESAs) are also sourced from this dataset. The challenge of accounting for ESAs are shown in KR2 section below.
GDP Controls: US Bureau of Economic Analysis. Values are also reported in current US dollars (real GDP values exist). The controls used in the below are total, private industry, and oil, gas, mining & quarrying county-level GDP.
Population controls: US Census Bureau.
Coal mine activity and production levels: Mine Safety and Health Administration
Descriptive Statistics
The below summary table displays descriptive statistics for all indicators used in regression results below.
All values for total enrollment, revenue, taxes, expenditure are in thousand USD. Values for GDP are reported in million USD. Values for population are reported in thousands.
GDP controls are currently only available at the county level from 2001-2021. Expanding this
% Table created by stargazer v.5.2.3 by Marek Hlavac, Social Policy Institute. E-mail: marek.hlavac at gmail.com % Date and time: Thu, Mar 21, 2024 - 11:26:58Descriptive Figures: County-level education expenditure
The following plot displays each county’s time series of expenditure (total and per pupil log values) as well as the overall mean. Colors correspond to the US state the county is in.
Key Relationships (KR)
Below are scatterplots and diagrams depicting key relationships between dependent and control variables as well as shares/components of key variables.
KR 1: Revenue Sources
Most important to tease out before modelling is how different revenue sources (federal, state, county (own), and other local sources) interplay. County-level revenue for public education is a combination of both local and intergovernmental sources. The local portion of the share is almost entirely sourced through property taxes. The intergovernmental sources come from state, federal, and local aid.
Below chart plots county-level mean (taken over the panel time horizon) of different intergovernmental (IG) sources versus own-source revenue (generated from local sources). The solid black line represents a best-fit line and the dashed line represents a 45 degree line. The blue plot shows Total IG Revenue (Federal + State + Other Local) versus own revenue. There is a near negative correlation between the two (ie. they are near-substitutes for one another). This effect is dominated by State IG revenue (as can be seen in the purple panel).
The following plots show the share of revenue for public education that comes from own sources, local intergovernmental, state intergovernmental, and federal intergovernmental by state. Almost all states have a near 50% split of revenue from state and own sources, which aligns with data from the Congressional Research Service cited in the Transfer of Status report. Massachusetts has an unusually high share of local IG support (inter-school aid), eclipsing own sources almost completely. From further research, I believe this has to do with a unique structure of Massachusetts public school funding which is reliant on several multi-county funding agencies (similar to the mentioned ESAs). This anomaly might warrant the exclusion of MA from the analysis.
Conclusion: I am still unsure how best to control for the effects of these shares on our results.
The below plot provides the same information as above but on a national level.
Conclusion: Corroborates the near-even split of revenue between state and own sources which seems to be a fact of public education revenue. I mainly include this as an alternative summarising figure to the above.
KR 2: ESAs and County Expenditure
Around 2007, many states instituted Educational Service Agencies (ESAs) which sought to “equalise” public education across the country. To date, ~45 states have ESAs which are responsible for multiple school districts, most often across multiple counties. Therefore, when modelling county-level expenditure it will be important to understand how this change in educational expenditure affected county-level spending (ie. did ESAs replace or supplement county-level funding).
Only 593 counties of 2740 in the dataset have recorded revenue/expenditure from ESAs. After some digging, I believe that these values for ESAs are improperly recorded in the sense that the revenue is recorded only in counties in which the ESA’s headquarters is located and not partitioned to the counties to which the revenue ultimately flows. All county-level total expenditure/revenue values that are used in the regressions on this page have explicitly excluded recorded ESA values for this reason (they are instead recorded in a variable called esa_tot_exp or esa_tot_rev in total and per pupil values).
The below graphs show some relationships between ESA and county-level finances. I have yet to arrive at a definitive understanding of how ESAs and county-level finances interact. They do not appear to be substitutes. As is evident, the data on ESA expenditure is patchy and highly volatile by county. At the moment, I believe this is because of imperfect/inconsistent reporting in comparison to traditional school district reporting. The four states that have recorded ESA expenditure before 2007 are California, Illinois, Minnesota, Oregon.
With this we want to demonstrate how ESAs interact with our expenditure indicator. Does high ESA spending imply low/high local spending? It seems from the below that the two have a no correlation, implying little substitution? (Each point is a county in a particular year starting from 2007, colour represents state). Warm color-scale is (top-left panel) is expenditure values whereas the cool color scales are revenue values. The black line is a 45 degree line.
KR 3: Revenue, Expenditure and Property Taxes
We know the majority of “own source” revenue comes from property taxes. The below scatterplots demonstrate the relationship between education revenue and expeniture versus property taxes collected. The solid black line represents a best-fit line and the dashed line represents a 45 degree line (intercept-adjusted to match data). The vertical dotted line represents a potential preliminary cutoff point for outliers - will be subject to further more rigorous testing.
KR 4: GDP and Property Taxes
The solid black line represents a best-fit line and the dashed line represents a 45 degree line (intercept-adjusted to match data).
GDP and Property taxes have a positive linear correlation, to be expected.
KR 5: GDP and Education Expenditure
The solid black line represents a best-fit line and the dashed line represents a 45 degree line (intercept-adjusted to match data).
GDP and Education expenditure have a positive linear correlation, as expected.
Preliminary Results
In the following set of results I report results for the sample used in the descriptive statistics above. With respect to the outlier threshold represented in KR3 I exclude any counties that report
Panel testing
is.pbalanced(df, index = c("fips", "year"))[1] TRUECounty-level trends
TBA: Summary of county-level trends to be added. Ideally graphics showing their spatial autocorrelation.
Descriptive regression models
The below models were run to test the relevance of various GDP control variables (total, private industry, oil, gas & mining) as well as corroborate some relationships in the above scatter plots of KRs. Ideally, once variables have been surveyed, they will be put through model selection via getspanel.
All regression models that follow include TWFE (county- and year- fixed effects) and standard errors clustered by county.
All functional forms in the feols() functions below are of the form “Y ~ X” In the cases in which multiple estimations are included via sw(Xa, Xb, Xc + Xd), the function will return results for “Y~Xa”, “Y~Xb”, “Y ~ Xc + Xd”.
Education Expenditure ~ GDP
A 1% increase in GDP is associated with a 0.14% increase in education expenditure, dominated by the effect of GDP from private industry (0.12%). I include here also the GDP generated from the oil, gas, mining, and quarrying sector. The effect is statistically significant but small.
# feols(c(log_Total_Educ_Total_Exp, log_Elem_Educ_Total_Exp) ~ sw(log_gdp_total,
# log_gdp_priv_ind,
# log_gdp_o_g_mining_quarr_21) | fips + year, data = mines_df, se = "cluster", cluster = "fips") %>% etable(se.below = TRUE) %>% kable
list(feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) + sw0(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Total_Educ_Total_Exp_pp ~ sw(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) | fips + year, data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)) %>% etable(se.below = TRUE) %>% kableNOTE: 2,710 observations removed because of NA values (RHS: 2,710).
|-> this msg only concerns the variables common to all estimations| .1 | .2 | .3 | .4 | .1 | .2 | .3 | |
|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp |
| l(log_Total_Educ_Total_Exp_pp,1) | 0.6058*** | 0.5844*** | 0.5869*** | 0.6034*** | |||
| (0.0089) | (0.0073) | (0.0074) | (0.0087) | ||||
| (log) GDP pc | 0.1010*** | 0.1994*** | |||||
| (0.0066) | (0.0136) | ||||||
| (log) GDP Priv. Industry pc | 0.0853*** | 0.1675*** | |||||
| (0.0059) | (0.0121) | ||||||
| (log) O&G&Mining pc | 0.0055*** | 0.0106*** | |||||
| (0.0007) | (0.0012) | ||||||
| Fixed-Effects: | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| ________________________________ | ___________________ | ___________________ | ___________________ | ___________________ | ___________________ | ___________________ | ___________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 54,200 | 54,200 | 54,200 | 54,200 | 56,700 | 56,700 | 56,700 |
| R2 | 0.87755 | 0.87992 | 0.87968 | 0.87783 | 0.82216 | 0.82107 | 0.81361 |
| Within R2 | 0.36193 | 0.37429 | 0.37305 | 0.36339 | 0.05131 | 0.04546 | 0.00569 |
list(feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) + sw0(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Elem_Educ_Total_Exp_pp ~ sw(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) | fips + year, data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)) %>% etable(se.below = TRUE) %>% kableNOTE: 2,710 observations removed because of NA values (RHS: 2,710).
|-> this msg only concerns the variables common to all estimations| .1 | .2 | .3 | .4 | .1 | .2 | .3 | |
|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| l(log_Elem_Educ_Total_Exp_pp,1) | 0.6041*** | 0.5827*** | 0.5852*** | 0.6017*** | |||
| (0.0090) | (0.0074) | (0.0075) | (0.0088) | ||||
| (log) GDP pc | 0.1007*** | 0.1984*** | |||||
| (0.0066) | (0.0136) | ||||||
| (log) GDP Priv. Industry pc | 0.0851*** | 0.1666*** | |||||
| (0.0059) | (0.0121) | ||||||
| (log) O&G&Mining pc | 0.0053*** | 0.0104*** | |||||
| (0.0007) | (0.0012) | ||||||
| Fixed-Effects: | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| _______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 54,200 | 54,200 | 54,200 | 54,200 | 56,700 | 56,700 | 56,700 |
| R2 | 0.87889 | 0.88124 | 0.88100 | 0.87915 | 0.82461 | 0.82352 | 0.81614 |
| Within R2 | 0.36002 | 0.37246 | 0.37121 | 0.36143 | 0.05139 | 0.04551 | 0.00556 |
Incorporating state-level trends
The below take the Education Expenditure ~ GDP models and incorporate state-level time trends.
list(feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) + sw0(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) +sw(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) + time:as.factor(state) | fips + year, data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)) %>% etable(., drop = "time*", se.below = TRUE) %>% kable| .1 | .2 | .3 | .4 | .1 | .2 | .3 | |
|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp |
| l(log_Total_Educ_Total_Exp_pp,1) | 0.5766*** | 0.5598*** | 0.5617*** | 0.5746*** | 0.5599*** | 0.5617*** | 0.5744*** |
| (0.0101) | (0.0082) | (0.0083) | (0.0099) | (0.0082) | (0.0083) | (0.0099) | |
| (log) GDP pc | 0.0948*** | 0.0943*** | |||||
| (0.0076) | (0.0076) | ||||||
| (log) GDP Priv. Industry pc | 0.0800*** | 0.0796*** | |||||
| (0.0067) | (0.0067) | ||||||
| (log) O&G&Mining pc | 0.0050*** | 0.0050*** | |||||
| (0.0007) | (0.0007) | ||||||
| Fixed-Effects: | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- | ------------------- |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| ________________________________ | ___________________ | ___________________ | ___________________ | ___________________ | ___________________ | ___________________ | ___________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 54,200 | 54,200 | 54,200 | 54,200 | 54,000 | 54,000 | 54,000 |
| R2 | 0.87974 | 0.88171 | 0.88152 | 0.87997 | 0.88102 | 0.88083 | 0.87930 |
| Within R2 | 0.37337 | 0.38360 | 0.38261 | 0.37456 | 0.38324 | 0.38226 | 0.37431 |
list(feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) + sw0(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) + sw(log_gdp_total_pc,
log_gdp_priv_ind_pc,
log_gdp_o_g_mining_quarr_21_pc) + time:as.factor(state) | fips + year, data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)) %>% etable(., drop = "time*", se.below = TRUE) %>% kable| .1 | .2 | .3 | .4 | .1 | .2 | .3 | |
|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| l(log_Elem_Educ_Total_Exp_pp,1) | 0.5750*** | 0.5581*** | 0.5600*** | 0.5731*** | 0.5581*** | 0.5600*** | 0.5729*** |
| (0.0102) | (0.0083) | (0.0084) | (0.0100) | (0.0083) | (0.0084) | (0.0101) | |
| (log) GDP pc | 0.0947*** | 0.0942*** | |||||
| (0.0076) | (0.0076) | ||||||
| (log) GDP Priv. Industry pc | 0.0799*** | 0.0795*** | |||||
| (0.0067) | (0.0067) | ||||||
| (log) O&G&Mining pc | 0.0048*** | 0.0048*** | |||||
| (0.0007) | (0.0007) | ||||||
| Fixed-Effects: | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| _______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 54,200 | 54,200 | 54,200 | 54,200 | 54,000 | 54,000 | 54,000 |
| R2 | 0.88105 | 0.88301 | 0.88281 | 0.88126 | 0.88233 | 0.88214 | 0.88060 |
| Within R2 | 0.37145 | 0.38179 | 0.38078 | 0.37259 | 0.38141 | 0.38041 | 0.37232 |
Incorporating time lags
Have not yet incorporated an AR term Education expenditure has a highly relevant time dependence. The effect of increases in GDP two years prior has the greatest effect on current education expenditure, implying a delayed effect of county-level economic prosperity/decline on public education expenditure.
Question: Is the above interpretation of the time dependence accurate?
levs_gdp_real_trend <- feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_total + log_gdp_total_l1 + log_gdp_total_l2, log_gdp_priv_ind + log_gdp_priv_ind_l1 + log_gdp_priv_ind_l2) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips")
levs_gdp_real_trend %>% etable(., drop = "time*", se.below = TRUE) %>% kable| ..1 | ..2 | ..3 | ..4 | |
|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP | 0.0044 | 0.0036 | ||
| (0.0109) | (0.0109) | |||
| (log,l1) GDP | 0.0578*** | 0.0582*** | ||
| (0.0077) | (0.0077) | |||
| (log,l2) GDP | 0.1268*** | 0.1272*** | ||
| (0.0176) | (0.0175) | |||
| (log) GDP Priv. Industry | 0.0041 | 0.0034 | ||
| (0.0094) | (0.0094) | |||
| (log,l1) GDP Priv. Industry | 0.0508*** | 0.0511*** | ||
| (0.0067) | (0.0066) | |||
| (log,l2) GDP Priv. Industry | 0.1101*** | 0.1104*** | ||
| (0.0151) | (0.0151) | |||
| Fixed-Effects: | ------------------- | ------------------- | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes |
| ___________________________ | ___________________ | ___________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.82611 | 0.82566 | 0.82845 | 0.82801 |
| Within R2 | 0.10747 | 0.10513 | 0.10761 | 0.10532 |
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2,
log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2,
log_gdp_o_g_mining_quarr_21 + log_gdp_o_g_mining_quarr_21_l1 + log_gdp_o_g_mining_quarr_21_l2) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips") %>% etable(., drop = "time*", se.below = TRUE) %>% kable| ..1 | ..2 | ..3 | ..4 | ..5 | ..6 | |
|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP pc | 0.0152 | 0.0141 | ||||
| (0.0109) | (0.0109) | |||||
| (log,l1) GDP pc | 0.0713*** | 0.0716*** | ||||
| (0.0104) | (0.0103) | |||||
| (log,l2) GDP pc | 0.1533*** | 0.1537*** | ||||
| (0.0164) | (0.0164) | |||||
| (log) GDP Priv. Industry pc | 0.0121 | 0.0112 | ||||
| (0.0093) | (0.0093) | |||||
| (log,l1) GDP Priv. Industry pc | 0.0614*** | 0.0615*** | ||||
| (0.0086) | (0.0086) | |||||
| (log,l2) GDP Priv. Industry pc | 0.1299*** | 0.1302*** | ||||
| (0.0143) | (0.0143) | |||||
| (log) GDP O&G&Mining | 0.0009** | 0.0009** | ||||
| (0.0003) | (0.0003) | |||||
| (log,l1) GDP O&G&Mining | 0.0007* | 0.0006 | ||||
| (0.0003) | (0.0003) | |||||
| (log,l2) GDP O&G&Mining | 0.0014*** | 0.0013*** | ||||
| (0.0004) | (0.0004) | |||||
| Fixed-Effects: | ------------------- | ------------------- | ------------------- | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes |
| ______________________________ | ___________________ | ___________________ | ___________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.82973 | 0.82867 | 0.81878 | 0.83203 | 0.83098 | 0.82108 |
| Within R2 | 0.12605 | 0.12059 | 0.06982 | 0.12621 | 0.12074 | 0.06927 |
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2,
log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2,
log_gdp_o_g_mining_quarr_21_pc + log_gdp_o_g_mining_quarr_21_pc_l1 + log_gdp_o_g_mining_quarr_21_pc_l2) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips") %>% etable(., drop = "time*", se.below = TRUE) %>% kable| ..1 | ..2 | ..3 | ..4 | ..5 | ..6 | |
|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP pc | 0.0152 | 0.0141 | ||||
| (0.0109) | (0.0109) | |||||
| (log,l1) GDP pc | 0.0713*** | 0.0716*** | ||||
| (0.0104) | (0.0103) | |||||
| (log,l2) GDP pc | 0.1533*** | 0.1537*** | ||||
| (0.0164) | (0.0164) | |||||
| (log) GDP Priv. Industry pc | 0.0121 | 0.0112 | ||||
| (0.0093) | (0.0093) | |||||
| (log,l1) GDP Priv. Industry pc | 0.0614*** | 0.0615*** | ||||
| (0.0086) | (0.0086) | |||||
| (log,l2) GDP Priv. Industry pc | 0.1299*** | 0.1302*** | ||||
| (0.0143) | (0.0143) | |||||
| (log) O&G&Mining pc | 0.0035** | 0.0034** | ||||
| (0.0011) | (0.0011) | |||||
| (log,l1) O&G&Mining pc | 0.0034** | 0.0033** | ||||
| (0.0012) | (0.0012) | |||||
| (log,l2) O&G&Mining pc | 0.0082*** | 0.0079*** | ||||
| (0.0014) | (0.0014) | |||||
| Fixed-Effects: | ------------------- | ------------------- | ------------------- | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes |
| ______________________________ | ___________________ | ___________________ | ___________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.82973 | 0.82867 | 0.81986 | 0.83203 | 0.83098 | 0.82212 |
| Within R2 | 0.12605 | 0.12059 | 0.07534 | 0.12621 | 0.12074 | 0.07465 |
Coal Mine & Production Data
Coal mine data exists from 1996-2021. The below is a preliminary treatment attempt using data on active coal production (total_active_prod) and active coal mines (total_active_n).
Question: might it be more interesting to look at the first difference of mining and production activity to understand the effect of a change in activity?
Standard TWFE panel regression
list(feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_n + l1_diff_total_active_n + l2_diff_total_active_n | fips + year, data = mines_df, se = "cluster", cluster = "fips"),
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_prod + l1_diff_total_active_prod + l2_diff_total_active_prod,
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + diff_total_active_n + l1_diff_total_active_n + l2_diff_total_active_n,
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + diff_total_active_prod + l1_diff_total_active_prod + l2_diff_total_active_prod) | fips + year, data = mines_df, se = "cluster", cluster = "fips")) %>% etable(se.below = TRUE) %>% kable| .1 | .2 | .1 | .2 | .3 | .4 | .5 | .6 | |
|---|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP Priv. Industry pc | 0.0145 | 0.0137 | 0.0145 | 0.0137 | ||||
| (0.0095) | (0.0095) | (0.0095) | (0.0095) | |||||
| (log,l1) GDP Priv. Industry pc | 0.0646*** | 0.0647*** | 0.0645*** | 0.0646*** | ||||
| (0.0086) | (0.0085) | (0.0086) | (0.0086) | |||||
| (log,l2) GDP Priv. Industry pc | 0.1462*** | 0.1457*** | 0.1463*** | 0.1457*** | ||||
| (0.0131) | (0.0131) | (0.0131) | (0.0131) | |||||
| (fd) Active Coal Mines | -0.0001 | -6.5e-5 | -0.0001 | -7.86e-5 | ||||
| (0.0011) | (0.0011) | (0.0011) | (0.0011) | |||||
| (l1, fd) Active Coal Mines | -0.0004 | -0.0004 | -0.0005 | -0.0004 | ||||
| (0.0010) | (0.0010) | (0.0010) | (0.0010) | |||||
| (l2, fd) Active Coal Mines | -0.0018* | -0.0017* | -0.0018* | -0.0018* | ||||
| (0.0009) | (0.0009) | (0.0009) | (0.0009) | |||||
| (fd) Active Coal Prod | -0.0003 | -0.0003 | -0.0003 | -0.0002 | ||||
| (0.0009) | (0.0009) | (0.0009) | (0.0009) | |||||
| (l1,fd) Active Coal Prod | -0.0005 | -0.0005 | -0.0005 | -0.0005 | ||||
| (0.0007) | (0.0007) | (0.0007) | (0.0007) | |||||
| (l2,fd) Active Coal Prod | -0.0019*** | -0.0019*** | -0.0018*** | -0.0019*** | ||||
| (0.0004) | (0.0004) | (0.0004) | (0.0004) | |||||
| (log) GDP pc | 0.0191 | 0.0191 | 0.0180 | 0.0180 | ||||
| (0.0112) | (0.0111) | (0.0111) | (0.0111) | |||||
| (log,l1) GDP pc | 0.0746*** | 0.0745*** | 0.0748*** | 0.0747*** | ||||
| (0.0103) | (0.0103) | (0.0103) | (0.0103) | |||||
| (log,l2) GDP pc | 0.1715*** | 0.1716*** | 0.1709*** | 0.1710*** | ||||
| (0.0150) | (0.0150) | (0.0150) | (0.0150) | |||||
| Fixed-Effects: | ------------------- | ------------------------------ | ------------------- | ------------------- | ------------------- | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| ______________________________ | ___________________ | ______________________________ | ___________________ | ___________________ | ___________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.81898 | 0.82141 | 0.81899 | 0.82041 | 0.82042 | 0.82142 | 0.82284 | 0.82284 |
| Within R2 | 0.07084 | 0.07099 | 0.07087 | 0.07820 | 0.07823 | 0.07102 | 0.07839 | 0.07842 |
list(feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(total_active_n, 0:2) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(total_active_prod, 0:2),
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + l(total_active_n, 0:2),
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + l(total_active_prod, 0:2)) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)) %>% etable(se.below = TRUE) %>% kable| .1 | .2 | .1 | .2 | .3 | .4 | .5 | .6 | |
|---|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP Priv. Industry pc | 0.0149 | 0.0140 | 0.0145 | 0.0137 | ||||
| (0.0095) | (0.0095) | (0.0095) | (0.0095) | |||||
| (log,l1) GDP Priv. Industry pc | 0.0646*** | 0.0647*** | 0.0645*** | 0.0646*** | ||||
| (0.0086) | (0.0086) | (0.0086) | (0.0086) | |||||
| (log,l2) GDP Priv. Industry pc | 0.1465*** | 0.1460*** | 0.1464*** | 0.1459*** | ||||
| (0.0131) | (0.0131) | (0.0132) | (0.0132) | |||||
| Active Coal Mines | -0.0005 | -0.0004 | -0.0006 | -0.0005 | ||||
| (0.0011) | (0.0011) | (0.0011) | (0.0011) | |||||
| (l1) Active Coal Mines | -0.0004 | -0.0004 | -0.0004 | -0.0004 | ||||
| (0.0009) | (0.0009) | (0.0009) | (0.0009) | |||||
| (l2) Active Coal Mines | -0.0009 | -0.0009 | -0.0009 | -0.0009 | ||||
| (0.0010) | (0.0010) | (0.0010) | (0.0010) | |||||
| Active Coal Prod | -1.56e-7 | -1.29e-7 | -1.26e-7 | -9.94e-8 | ||||
| (8.32e-7) | (8.31e-7) | (8.49e-7) | (8.48e-7) | |||||
| (l1) Active Coal Prod | -2.9e-7 | -3.08e-7 | -2.91e-7 | -3.1e-7 | ||||
| (3.45e-7) | (3.43e-7) | (3.5e-7) | (3.48e-7) | |||||
| (l2) Active Coal Prod | 3.5e-8 | -9.07e-8 | 5.6e-8 | -6.98e-8 | ||||
| (1.45e-6) | (1.47e-6) | (1.44e-6) | (1.46e-6) | |||||
| (log) GDP pc | 0.0195 | 0.0191 | 0.0184 | 0.0180 | ||||
| (0.0112) | (0.0111) | (0.0111) | (0.0111) | |||||
| (log,l1) GDP pc | 0.0746*** | 0.0745*** | 0.0748*** | 0.0747*** | ||||
| (0.0103) | (0.0103) | (0.0103) | (0.0103) | |||||
| (log,l2) GDP pc | 0.1719*** | 0.1717*** | 0.1712*** | 0.1711*** | ||||
| (0.0150) | (0.0150) | (0.0150) | (0.0150) | |||||
| Fixed-Effects: | ------------------- | ------------------------------ | ------------------- | ------------------- | ------------------- | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| ______________________________ | ___________________ | ______________________________ | ___________________ | ___________________ | ___________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.81899 | 0.82142 | 0.81897 | 0.82043 | 0.82041 | 0.82141 | 0.82285 | 0.82283 |
| Within R2 | 0.07091 | 0.07104 | 0.07082 | 0.07828 | 0.07818 | 0.07096 | 0.07846 | 0.07837 |
TWFE panel with state-level trends
list(feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_n + l1_diff_total_active_n + l2_diff_total_active_n + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips"),
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_prod + l1_diff_total_active_prod + l2_diff_total_active_prod,
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + diff_total_active_n + l1_diff_total_active_n + l2_diff_total_active_n,
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + diff_total_active_prod + l1_diff_total_active_prod + l2_diff_total_active_prod) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips")) %>% etable(se.below = TRUE, drop = "time*") %>% kable| .1 | .2 | .1 | .2 | .3 | .4 | .5 | .6 | |
|---|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP Priv. Industry pc | 0.0123 | 0.0114 | 0.0123 | 0.0113 | ||||
| (0.0093) | (0.0093) | (0.0093) | (0.0093) | |||||
| (log,l1) GDP Priv. Industry pc | 0.0615*** | 0.0617*** | 0.0615*** | 0.0616*** | ||||
| (0.0086) | (0.0086) | (0.0087) | (0.0086) | |||||
| (log,l2) GDP Priv. Industry pc | 0.1297*** | 0.1300*** | 0.1298*** | 0.1301*** | ||||
| (0.0142) | (0.0143) | (0.0142) | (0.0143) | |||||
| (fd) Active Coal Mines | -0.0003 | -0.0003 | -0.0003 | -0.0003 | ||||
| (0.0010) | (0.0010) | (0.0010) | (0.0010) | |||||
| (l1, fd) Active Coal Mines | -0.0005 | -0.0005 | -0.0005 | -0.0005 | ||||
| (0.0009) | (0.0009) | (0.0009) | (0.0009) | |||||
| (l2, fd) Active Coal Mines | -0.0020* | -0.0019* | -0.0020* | -0.0020* | ||||
| (0.0008) | (0.0008) | (0.0008) | (0.0008) | |||||
| (fd) Active Coal Prod | -0.0004 | -0.0004 | -0.0004 | -0.0004 | ||||
| (0.0007) | (0.0007) | (0.0007) | (0.0007) | |||||
| (l1,fd) Active Coal Prod | -0.0006 | -0.0006 | -0.0006 | -0.0006 | ||||
| (0.0006) | (0.0006) | (0.0006) | (0.0006) | |||||
| (l2,fd) Active Coal Prod | -0.0019*** | -0.0020*** | -0.0019*** | -0.0020*** | ||||
| (0.0005) | (0.0005) | (0.0005) | (0.0005) | |||||
| (log) GDP pc | 0.0154 | 0.0154 | 0.0143 | 0.0143 | ||||
| (0.0109) | (0.0109) | (0.0109) | (0.0109) | |||||
| (log,l1) GDP pc | 0.0715*** | 0.0714*** | 0.0717*** | 0.0717*** | ||||
| (0.0104) | (0.0104) | (0.0104) | (0.0104) | |||||
| (log,l2) GDP pc | 0.1531*** | 0.1532*** | 0.1534*** | 0.1535*** | ||||
| (0.0164) | (0.0165) | (0.0164) | (0.0165) | |||||
| Fixed-Effects: | ------------------- | ------------------------------ | ------------------- | ------------------- | ------------------- | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| ______________________________ | ___________________ | ______________________________ | ___________________ | ___________________ | ___________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.82869 | 0.83099 | 0.82868 | 0.82974 | 0.82974 | 0.83099 | 0.83203 | 0.83205 |
| Within R2 | 0.12069 | 0.12081 | 0.12066 | 0.12608 | 0.12609 | 0.12080 | 0.12622 | 0.12630 |
list(feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(total_active_n, 0:2) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ sw(log_gdp_priv_ind_pc + log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(total_active_prod, 0:2),
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + l(total_active_n, 0:2),
log_gdp_total_pc + log_gdp_total_pc_l1 + log_gdp_total_pc_l2 + l(total_active_prod, 0:2)) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)) %>% etable(se.below = TRUE, drop = "time*") %>% kable| .1 | .2 | .1 | .2 | .3 | .4 | .5 | .6 | |
|---|---|---|---|---|---|---|---|---|
| Dependent Var.: | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp | (log) Elementary Educ. Exp. pp |
| (log) GDP Priv. Industry pc | 0.0127 | 0.0118 | 0.0123 | 0.0114 | ||||
| (0.0094) | (0.0093) | (0.0093) | (0.0093) | |||||
| (log,l1) GDP Priv. Industry pc | 0.0616*** | 0.0617*** | 0.0614*** | 0.0616*** | ||||
| (0.0087) | (0.0087) | (0.0087) | (0.0087) | |||||
| (log,l2) GDP Priv. Industry pc | 0.1301*** | 0.1303*** | 0.1299*** | 0.1302*** | ||||
| (0.0143) | (0.0143) | (0.0143) | (0.0143) | |||||
| Active Coal Mines | -0.0008 | -0.0008 | -0.0008 | -0.0008 | ||||
| (0.0010) | (0.0010) | (0.0010) | (0.0010) | |||||
| (l1) Active Coal Mines | -0.0002 | -0.0002 | -0.0002 | -0.0002 | ||||
| (0.0009) | (0.0009) | (0.0009) | (0.0009) | |||||
| (l2) Active Coal Mines | -0.0010 | -0.0009 | -0.0010 | -0.0010 | ||||
| (0.0010) | (0.0010) | (0.0009) | (0.0009) | |||||
| Active Coal Prod | -2.88e-7 | -2.8e-7 | -2.82e-7 | -2.73e-7 | ||||
| (6.57e-7) | (6.56e-7) | (6.68e-7) | (6.88e-7) | |||||
| (l1) Active Coal Prod | -2.04e-7 | -2.3e-7 | -2.06e-7 | -2.32e-7 | ||||
| (3.22e-7) | (3.22e-7) | (3.25e-7) | (3.22e-7) | |||||
| (l2) Active Coal Prod | 9.17e-8 | -5.09e-9 | 9.66e-8 | -7.66e-10 | ||||
| (1.38e-6) | (1.39e-6) | (1.38e-6) | (1.4e-6) | |||||
| (log) GDP pc | 0.0159 | 0.0154 | 0.0147 | 0.0143 | ||||
| (0.0109) | (0.0109) | (0.0109) | (0.0109) | |||||
| (log,l1) GDP pc | 0.0715*** | 0.0714*** | 0.0717*** | 0.0716*** | ||||
| (0.0104) | (0.0104) | (0.0104) | (0.0104) | |||||
| (log,l2) GDP pc | 0.1535*** | 0.1534*** | 0.1538*** | 0.1537*** | ||||
| (0.0164) | (0.0164) | (0.0165) | (0.0164) | |||||
| Fixed-Effects: | ------------------- | ------------------------------ | ------------------- | ------------------- | ------------------- | ------------------------------ | ------------------------------ | ------------------------------ |
| fips | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| year | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
| ______________________________ | ___________________ | ______________________________ | ___________________ | ___________________ | ___________________ | ______________________________ | ______________________________ | ______________________________ |
| S.E.: Clustered | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips | by: fips |
| Observations | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 | 51,490 |
| R2 | 0.82869 | 0.83100 | 0.82868 | 0.82976 | 0.82974 | 0.83098 | 0.83206 | 0.83203 |
| Within R2 | 0.12070 | 0.12083 | 0.12065 | 0.12616 | 0.12605 | 0.12076 | 0.12635 | 0.12620 |
Dealing with endogeneity
There is a significant endogeneity concern in using total active production and active mines as the treatment variable. Therefore, I have tried two instrumental variable approaches below and aim to add results using production- and employment-based Bartik instruments.
Instrumental Variable
We consider using total active production or total active mines as an instrument affecting education expenditure through property taxes or GDP. We know that property taxes have an endogenous relationship with education expenditure (it is a component of the public education budget), however, in theory mine closures/activity is unlikely to affect education expenditure, except via property taxes. We test this hypothesis below.
Conclusion: The results themselves are not promising (I think??)
Using coal production and mine activity as instruments for per capita property tax
To my eye, I don’t see any of these being useful - either they fail a first-stage F-test or the resulting coefficient on the treatment is statistically insignificant.
# Using total active production as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ total_active_prod, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: total_active_prod
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 54,200
Fixed-effects: fips: 2,710, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 0.380481 0.161552 2.35516 1.8586e-02 *
l(log_Total_Educ_Total_Exp_pp, 1) 0.418648 0.079832 5.24413 1.6914e-07 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.127886 Adj. R2: 0.837506
Within R2: 0.195924
F-test (1st stage), (log) Property Taxes pp: stat = 6.99356, p = 0.008183, on 1 and 54,178 DoF.
Wu-Hausman: stat = 2.19171, p = 0.138761, on 1 and 51,468 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ total_active_prod, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: total_active_prod
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 54,200
Fixed-effects: fips: 2,710, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 0.391712 0.166631 2.35077 1.8806e-02 *
l(log_Elem_Educ_Total_Exp_pp, 1) 0.412008 0.082105 5.01804 5.5581e-07 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.129112 Adj. R2: 0.834624
Within R2: 0.170155
F-test (1st stage), (log) Property Taxes pp: stat = 6.7651 , p = 0.009298, on 1 and 54,178 DoF.
Wu-Hausman: stat = 2.35562, p = 0.124838, on 1 and 51,468 DoF.# Using the lag of total active production as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ l(total_active_prod,1), data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: l(total_active_prod, 1)
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 54,000
Fixed-effects: fips: 2,700, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 0.410549 0.219347 1.87169 0.06135785 .
l(log_Total_Educ_Total_Exp_pp, 1) 0.405674 0.107155 3.78586 0.00015651 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.119107 Adj. R2: 0.858302
Within R2: 0.302456
F-test (1st stage), (log) Property Taxes pp: stat = 11.0 , p = 9.151e-4, on 1 and 53,978 DoF.
Wu-Hausman: stat = 1.94096, p = 0.16357 , on 1 and 51,278 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ l(total_active_prod,1), data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: l(total_active_prod, 1)
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 54,000
Fixed-effects: fips: 2,700, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 0.421586 0.225726 1.86769 0.06191303 .
l(log_Elem_Educ_Total_Exp_pp, 1) 0.399242 0.109927 3.63187 0.00028661 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.119875 Adj. R2: 0.856682
Within R2: 0.284533
F-test (1st stage), (log) Property Taxes pp: stat = 10.6 , p = 0.001115, on 1 and 53,978 DoF.
Wu-Hausman: stat = 2.12024, p = 0.14537 , on 1 and 51,278 DoF.# Using total active mines as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ total_active_n, data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: total_active_n
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 54,000
Fixed-effects: fips: 2,700, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 1.103790 1.132518 0.974634 0.32983
l(log_Total_Educ_Total_Exp_pp, 1) 0.067935 0.551994 0.123072 0.90206
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.211179 Adj. R2: 0.554561
Within R2: -1.19279
F-test (1st stage), (log) Property Taxes pp: stat = 1.71329, p = 0.190564, on 1 and 53,978 DoF.
Wu-Hausman: stat = 4.44018, p = 0.035107, on 1 and 51,278 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ total_active_n, data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: total_active_n
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 54,000
Fixed-effects: fips: 2,700, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 1.269797 1.484555 0.855339 0.39244
l(log_Elem_Educ_Total_Exp_pp, 1) -0.012671 0.721284 -0.017567 0.98599
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.239677 Adj. R2: 0.427071
Within R2: -1.86014
F-test (1st stage), (log) Property Taxes pp: stat = 1.29917, p = 0.254369, on 1 and 53,978 DoF.
Wu-Hausman: stat = 4.73616, p = 0.02954 , on 1 and 51,278 DoF.# Using lag of total active mines as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ l(log_Total_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ l(total_active_n,1:2), data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: l(total_active_n, 1), l(total_active_n, 2)
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 51,300
Fixed-effects: fips: 2,700, year: 19
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 0.462123 0.238564 1.93711 0.0528365 .
l(log_Total_Educ_Total_Exp_pp, 1) 0.375618 0.113156 3.31946 0.0009139 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.123169 Adj. R2: 0.843794
Within R2: 0.244667
F-test (1st stage), (log) Property Taxes pp: stat = 3.51732 , p = 0.029686, on 2 and 51,278 DoF.
Wu-Hausman: stat = 1.75648 , p = 0.185071, on 1 and 48,579 DoF.
Sargan: stat = 0.655074, p = 0.418305, on 1 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ l(log_Elem_Educ_Total_Exp_pp, 1) | fips + year | log_Property_Tax_pp ~ l(total_active_n,1:2), data = mines_restr, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_Property_Tax_pp, Instr.: l(total_active_n, 1), l(total_active_n, 2)
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 51,300
Fixed-effects: fips: 2,700, year: 19
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_Property_Tax_pp 0.462834 0.248439 1.86297 0.0625756 .
l(log_Elem_Educ_Total_Exp_pp, 1) 0.374285 0.117442 3.18698 0.0014541 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.123211 Adj. R2: 0.843942
Within R2: 0.235191
F-test (1st stage), (log) Property Taxes pp: stat = 3.27467 , p = 0.037837, on 2 and 51,278 DoF.
Wu-Hausman: stat = 1.69032 , p = 0.193565, on 1 and 48,579 DoF.
Sargan: stat = 0.675589, p = 0.41111 , on 1 DoF.Using coal production and mine activity as instruments for per capita gdp
All pass a first-stage F-test with high statistic and low p-value; however, less than satisfactory on the Wu-Hausman endogeneity test.
It seems that if we replace the “total education expenditure per pupil” with a measure of total expenditure on elementary education only, the effect is more pronounced.
Conclusion: This is the results I am most inclined to “report” in some way, even though it is weak. So I am curious for your thoughts on the validity (and blind spots I’ve missed when constructing the model).
# Using total active production as an instrument for per capita gdp
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ total_active_prod, data = mines_df, se = "cluster", cluster = "fips")TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: total_active_prod
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 56,910
Fixed-effects: fips: 2,710, year: 21
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.133674 0.032813 4.07388 4.7565e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.139391 Adj. R2: 0.81311
Within R2: 0.04608
F-test (1st stage), (log) GDP pc: stat = 272.9 , p < 2.2e-16 , on 1 and 56,888 DoF.
Wu-Hausman: stat = 1.56935, p = 0.210307, on 1 and 54,178 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ total_active_prod, data = mines_df, se = "cluster", cluster = "fips")TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: total_active_prod
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 56,910
Fixed-effects: fips: 2,710, year: 21
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.141252 0.032399 4.35971 1.3509e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.138426 Adj. R2: 0.815934
Within R2: 0.047508
F-test (1st stage), (log) GDP pc: stat = 272.9 , p < 2.2e-16 , on 1 and 56,888 DoF.
Wu-Hausman: stat = 1.20347, p = 0.272635, on 1 and 54,178 DoF.# Using the lag of total active production as an instrument for per capita gdp
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ l(total_active_prod,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: l(total_active_prod, 1)
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 54,200
Fixed-effects: fips: 2,710, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.175387 0.043746 4.00923 6.2562e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.139101 Adj. R2: 0.807762
Within R2: 0.048723
F-test (1st stage), (log) GDP pc: stat = 260.2 , p < 2.2e-16 , on 1 and 54,179 DoF.
Wu-Hausman: stat = 0.172179, p = 0.678184, on 1 and 51,469 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ l(total_active_prod,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: l(total_active_prod, 1)
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 54,200
Fixed-effects: fips: 2,710, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.181331 0.043199 4.19758 2.7847e-05 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.138211 Adj. R2: 0.810498
Within R2: 0.049074
F-test (1st stage), (log) GDP pc: stat = 260.2 , p < 2.2e-16 , on 1 and 54,179 DoF.
Wu-Hausman: stat = 0.082524, p = 0.773907, on 1 and 51,469 DoF.# Using total active mines as an instrument for per capita gdp
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ total_active_n, data = mines_df, se = "cluster", cluster = "fips")TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: total_active_n
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 56,910
Fixed-effects: fips: 2,710, year: 21
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.087459 0.059381 1.47286 0.1409
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.140172 Adj. R2: 0.811012
Within R2: 0.035371
F-test (1st stage), (log) GDP pc: stat = 374.9 , p < 2.2e-16 , on 1 and 56,888 DoF.
Wu-Hausman: stat = 6.18839, p = 0.012862, on 1 and 54,178 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ total_active_n, data = mines_df, se = "cluster", cluster = "fips")TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: total_active_n
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 56,910
Fixed-effects: fips: 2,710, year: 21
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.106217 0.055931 1.89907 0.057662 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.138927 Adj. R2: 0.814597
Within R2: 0.040593
F-test (1st stage), (log) GDP pc: stat = 374.9 , p < 2.2e-16 , on 1 and 56,888 DoF.
Wu-Hausman: stat = 4.2574, p = 0.039084, on 1 and 54,178 DoF.# Using lag og total active mines as an instrument for per capita gdp
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ l(total_active_n,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Total_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: l(total_active_n, 1)
Second stage: Dep. Var.: log_Total_Educ_Total_Exp_pp
Observations: 54,200
Fixed-effects: fips: 2,710, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.111891 0.058146 1.92432 0.05442 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.139737 Adj. R2: 0.805999
Within R2: 0.040002
F-test (1st stage), (log) GDP pc: stat = 359.5 , p < 2.2e-16 , on 1 and 54,179 DoF.
Wu-Hausman: stat = 3.37167, p = 0.066332, on 1 and 51,469 DoF.feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_gdp_total_pc ~ l(total_active_n,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)TSLS estimation, Dep. Var.: log_Elem_Educ_Total_Exp_pp, Endo.: log_gdp_total_pc, Instr.: l(total_active_n, 1)
Second stage: Dep. Var.: log_Elem_Educ_Total_Exp_pp
Observations: 54,200
Fixed-effects: fips: 2,710, year: 20
Standard-errors: Clustered (fips)
Estimate Std. Error t value Pr(>|t|)
fit_log_gdp_total_pc 0.124532 0.056655 2.19806 0.028029 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
RMSE: 0.138674 Adj. R2: 0.809226
Within R2: 0.042692
F-test (1st stage), (log) GDP pc: stat = 359.5 , p < 2.2e-16 , on 1 and 54,179 DoF.
Wu-Hausman: stat = 2.40697, p = 0.120802, on 1 and 51,469 DoF.Regression tables for ToS
Standard controls and coal production data
# Total educ w/0 county-level trends
feols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ log_gdp_priv_ind_pc +
sw0(log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2,
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_n,
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(diff_total_active_n, 0:2),
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_prod,
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(diff_total_active_prod, 0:2)) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year) %>% etable(se.below = TRUE, tex = TRUE)\begingroup
\centering
\begin{tabular}{lcccccccccccc}
\tabularnewline \midrule \midrule
Dependent Variables: & \multicolumn{6}{c}{(log) Educ. Exp. pp} & \multicolumn{6}{c}{(log) Elementary Educ. Exp. pp}\\
Model: & (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) & (10) & (11) & (12)\\
\midrule
\emph{Variables}\\
(log) GDP Priv. Industry pc & 0.1683$^{***}$ & 0.0143 & 0.0143 & 0.0145 & 0.0144 & 0.0145 & 0.1674$^{***}$ & 0.0135 & 0.0135 & 0.0137 & 0.0135 & 0.0137\\
& (0.0121) & (0.0095) & (0.0095) & (0.0095) & (0.0095) & (0.0095) & (0.0121) & (0.0095) & (0.0095) & (0.0095) & (0.0095) & (0.0095)\\
(log,l1) GDP Priv. Industry pc & & 0.0644$^{***}$ & 0.0644$^{***}$ & 0.0646$^{***}$ & 0.0644$^{***}$ & 0.0645$^{***}$ & & 0.0645$^{***}$ & 0.0645$^{***}$ & 0.0647$^{***}$ & 0.0645$^{***}$ & 0.0646$^{***}$\\
& & (0.0085) & (0.0085) & (0.0086) & (0.0086) & (0.0086) & & (0.0085) & (0.0085) & (0.0085) & (0.0085) & (0.0086)\\
(log,l2) GDP Priv. Industry pc & & 0.1464$^{***}$ & 0.1464$^{***}$ & 0.1462$^{***}$ & 0.1464$^{***}$ & 0.1463$^{***}$ & & 0.1459$^{***}$ & 0.1459$^{***}$ & 0.1457$^{***}$ & 0.1459$^{***}$ & 0.1457$^{***}$\\
& & (0.0131) & (0.0131) & (0.0131) & (0.0131) & (0.0131) & & (0.0131) & (0.0131) & (0.0131) & (0.0131) & (0.0131)\\
(fd) Active Coal Mines & & & $-2.53\times 10^{-5}$ & -0.0001 & & & & & $2.78\times 10^{-5}$ & $-6.5\times 10^{-5}$ & & \\
& & & (0.0010) & (0.0011) & & & & & (0.0010) & (0.0011) & & \\
(l1,fd) Active Coal Mines & & & & -0.0004 & & & & & & -0.0004 & & \\
& & & & (0.0010) & & & & & & (0.0010) & & \\
(l2,fd) Active Coal Mines & & & & -0.0018$^{**}$ & & & & & & -0.0017$^{**}$ & & \\
& & & & (0.0009) & & & & & & (0.0009) & & \\
(fd) Active Coal Prod & & & & & $-9.32\times 10^{-5}$ & -0.0003 & & & & & $-6.88\times 10^{-5}$ & -0.0003\\
& & & & & (0.0008) & (0.0009) & & & & & (0.0008) & (0.0009)\\
(l1,fd) Active Coal Prod & & & & & & -0.0005 & & & & & & -0.0005\\
& & & & & & (0.0007) & & & & & & (0.0007)\\
(l2,fd) Active Coal Prod & & & & & & -0.0019$^{***}$ & & & & & & -0.0018$^{***}$\\
& & & & & & (0.0004) & & & & & & (0.0004)\\
\midrule
\emph{Fixed-effects}\\
fips & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes\\
year & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes\\
\midrule
\emph{Fit statistics}\\
Observations & 56,910 & 51,490 & 51,490 & 51,490 & 51,490 & 51,490 & 56,910 & 51,490 & 51,490 & 51,490 & 51,490 & 51,490\\
R$^2$ & 0.82205 & 0.81897 & 0.81897 & 0.81898 & 0.81897 & 0.81899 & 0.82448 & 0.82141 & 0.82141 & 0.82141 & 0.82141 & 0.82142\\
Within R$^2$ & 0.04592 & 0.07081 & 0.07081 & 0.07084 & 0.07081 & 0.07087 & 0.04598 & 0.07095 & 0.07095 & 0.07099 & 0.07095 & 0.07102\\
\midrule \midrule
\multicolumn{13}{l}{\emph{Clustered (fips) standard-errors in parentheses}}\\
\multicolumn{13}{l}{\emph{Signif. Codes: ***: 0.01, **: 0.05, *: 0.1}}\\
\end{tabular}
\par\endgroupfeols(c(log_Total_Educ_Total_Exp_pp, log_Elem_Educ_Total_Exp_pp) ~ log_gdp_priv_ind_pc +
sw0(log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2,
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_n,
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(diff_total_active_n, 0:2),
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + diff_total_active_prod,
log_gdp_priv_ind_pc_l1 + log_gdp_priv_ind_pc_l2 + l(diff_total_active_prod, 0:2)) + time:as.factor(state) | fips + year, data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year) %>% etable(se.below = TRUE, drop = "time*", tex = TRUE)Notes from the estimations:
[x 2] The variable 'time:as.factor(state)56' has been removed because of collinearity (see $collin.var).\begingroup
\centering
\begin{tabular}{lcccccccccccc}
\tabularnewline \midrule \midrule
Dependent Variables: & \multicolumn{6}{c}{(log) Educ. Exp. pp} & \multicolumn{6}{c}{(log) Elementary Educ. Exp. pp}\\
Model: & (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8) & (9) & (10) & (11) & (12)\\
\midrule
\emph{Variables}\\
(log) GDP Priv. Industry pc & 0.1446$^{***}$ & 0.0121 & 0.0121 & 0.0123 & 0.0122 & 0.0123 & 0.1441$^{***}$ & 0.0112 & 0.0112 & 0.0114 & 0.0112 & 0.0113\\
& (0.0135) & (0.0093) & (0.0093) & (0.0093) & (0.0093) & (0.0093) & (0.0135) & (0.0093) & (0.0093) & (0.0093) & (0.0093) & (0.0093)\\
(log,l1) GDP Priv. Industry pc & & 0.0614$^{***}$ & 0.0614$^{***}$ & 0.0615$^{***}$ & 0.0614$^{***}$ & 0.0615$^{***}$ & & 0.0615$^{***}$ & 0.0615$^{***}$ & 0.0617$^{***}$ & 0.0615$^{***}$ & 0.0616$^{***}$\\
& & (0.0086) & (0.0086) & (0.0086) & (0.0086) & (0.0087) & & (0.0086) & (0.0086) & (0.0086) & (0.0086) & (0.0086)\\
(log,l2) GDP Priv. Industry pc & & 0.1299$^{***}$ & 0.1299$^{***}$ & 0.1297$^{***}$ & 0.1299$^{***}$ & 0.1298$^{***}$ & & 0.1302$^{***}$ & 0.1302$^{***}$ & 0.1300$^{***}$ & 0.1302$^{***}$ & 0.1301$^{***}$\\
& & (0.0143) & (0.0142) & (0.0142) & (0.0142) & (0.0142) & & (0.0143) & (0.0143) & (0.0143) & (0.0142) & (0.0143)\\
(fd) Active Coal Mines & & & -0.0002 & -0.0003 & & & & & -0.0002 & -0.0003 & & \\
& & & (0.0009) & (0.0010) & & & & & (0.0009) & (0.0010) & & \\
(l1,fd) Active Coal Mines & & & & -0.0005 & & & & & & -0.0005 & & \\
& & & & (0.0009) & & & & & & (0.0009) & & \\
(l2,fd) Active Coal Mines & & & & -0.0020$^{**}$ & & & & & & -0.0019$^{**}$ & & \\
& & & & (0.0008) & & & & & & (0.0008) & & \\
(fd) Active Coal Prod & & & & & -0.0002 & -0.0004 & & & & & -0.0002 & -0.0004\\
& & & & & (0.0006) & (0.0007) & & & & & (0.0007) & (0.0007)\\
(l1,fd) Active Coal Prod & & & & & & -0.0006 & & & & & & -0.0006\\
& & & & & & (0.0006) & & & & & & (0.0006)\\
(l2,fd) Active Coal Prod & & & & & & -0.0019$^{***}$ & & & & & & -0.0019$^{***}$\\
& & & & & & (0.0005) & & & & & & (0.0005)\\
\midrule
\emph{Fixed-effects}\\
fips & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes\\
year & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes & Yes\\
\midrule
\emph{Fit statistics}\\
Observations & 56,910 & 51,490 & 51,490 & 51,490 & 51,490 & 51,490 & 56,910 & 51,490 & 51,490 & 51,490 & 51,490 & 51,490\\
R$^2$ & 0.83389 & 0.82867 & 0.82868 & 0.82869 & 0.82867 & 0.82868 & 0.83612 & 0.83098 & 0.83098 & 0.83099 & 0.83097 & 0.83099\\
Within R$^2$ & 0.10943 & 0.12059 & 0.12063 & 0.12069 & 0.12060 & 0.12066 & 0.10921 & 0.12074 & 0.12074 & 0.12081 & 0.12072 & 0.12080\\
\midrule \midrule
\multicolumn{13}{l}{\emph{Clustered (fips) standard-errors in parentheses}}\\
\multicolumn{13}{l}{\emph{Signif. Codes: ***: 0.01, **: 0.05, *: 0.1}}\\
\end{tabular}
\par\endgroupInstrumental Variable
setFixest_etable(fitstat = ~ . + ivf1 + ivf1.p + wh + wh.p + ivwald1 + ivwald1.p + ivf2 + ivf2.p + ivwald2 + ivwald2.p, se.below = TRUE)
# Using total active production as an instrument for per capita property tax
lapply(list(feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ total_active_prod, data = mines_df, se = "cluster", cluster = "fips"),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ total_active_prod, data = mines_df, se = "cluster", cluster = "fips"),
# Using the lag of total active production as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(total_active_prod,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(total_active_prod,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)), summary) %>% etable(tex = TRUE)NOTE: 2,710 observations removed because of NA values (IV: 0/2,710).
NOTE: 2,710 observations removed because of NA values (IV: 0/2,710).\begingroup
\centering
\begin{tabular}{lcccc}
\tabularnewline \midrule \midrule
Dependent Variables: & (log) Educ. Exp. pp & (log) Elementary Educ. Exp. pp & (log) Educ. Exp. pp & (log) Elementary Educ. Exp. pp\\
Model: & (1) & (2) & (3) & (4)\\
\midrule
\emph{Variables}\\
(log) Property Taxes pp & 0.4690$^{***}$ & 0.4956$^{***}$ & 0.5674$^{**}$ & 0.5866$^{***}$\\
& (0.1572) & (0.1598) & (0.2240) & (0.2260)\\
\midrule
\emph{Fixed-effects}\\
fips & Yes & Yes & Yes & Yes\\
year & Yes & Yes & Yes & Yes\\
\midrule
\emph{Fit statistics}\\
Observations & 56,910 & 56,910 & 54,200 & 54,200\\
R$^2$ & 0.78874 & 0.78018 & 0.75413 & 0.74631\\
Within R$^2$ & -0.13264 & -0.19484 & -0.28119 & -0.34051\\
F-test (1st stage), (log) Property Taxes pp & 8.9340 & 8.9340 & 10.439 & 10.439\\
F-test (1st stage), p-value, (log) Property Taxes pp & 0.00280 & 0.00280 & 0.00123 & 0.00123\\
Wu-Hausman & 2.3866 & 2.9526 & 4.5142 & 5.1377\\
Wu-Hausman, p-value & 0.12238 & 0.08575 & 0.03362 & 0.02342\\
Wald (1st stage), (log) Property Taxes pp & 0.05886 & 0.05886 & 0.07679 & 0.07679\\
Wald (1st stage), p-value, (log) Property Taxes pp & 0.80831 & 0.80831 & 0.78170 & 0.78170\\
F-test (2nd stage) & 6.2720 & 7.0908 & 10.001 & 10.825\\
F-test (2nd stage), p-value & 0.01227 & 0.00775 & 0.00157 & 0.00100\\
Wald (2nd stage) & 8.9015 & 9.6151 & 6.4189 & 6.7377\\
Wald (2nd stage), p-value & 0.00285 & 0.00193 & 0.01129 & 0.00944\\
\midrule \midrule
\multicolumn{5}{l}{\emph{Clustered (fips) standard-errors in parentheses}}\\
\multicolumn{5}{l}{\emph{Signif. Codes: ***: 0.01, **: 0.05, *: 0.1}}\\
\end{tabular}
\par\endgroup# Using total active mines as an instrument for per capita property tax
lapply(list(feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ total_active_n, data = mines_df, se = "cluster", cluster = "fips"),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ total_active_n, data = mines_df, se = "cluster", cluster = "fips"),
# Using lag of total active mines as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(total_active_n,1:2), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(total_active_n,1:2), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)), summary) %>% etable()NOTE: 5,420 observations removed because of NA values (IV: 0/5,420).NOTE: 5,420 observations removed because of NA values (IV: 0/5,420). model 1
Dependent Var.: (log) Educ. Exp. pp
(log) Property Taxes pp 1.111
(1.190)
Fixed-Effects: -------------------
fips Yes
year Yes
________________________________________ ___________________
S.E.: Clustered by: fips
Observations 56,910
R2 0.33058
Within R2 -2.5890
F-test (1st stage), (log) Property Taxes pp 0.93481
F-test (1st stage), p-value, (log) Property Taxes pp 0.33362
Wu-Hausman 2.7221
Wu-Hausman, p-value 0.09897
Wald (1st stage), (log) Property Taxes pp 0.73637
Wald (1st stage), p-value, (log) Property Taxes pp 0.39083
F-test (2nd stage) 3.6810
F-test (2nd stage), p-value 0.05504
Wald (2nd stage) 0.87107
Wald (2nd stage), p-value 0.35066
model 2
Dependent Var.: (log) Elementary Educ. Exp. pp
(log) Property Taxes pp 1.349
(1.403)
Fixed-Effects: ------------------------------
fips Yes
year Yes
________________________________________ ______________________________
S.E.: Clustered by: fips
Observations 56,910
R2 0.03410
Within R2 -4.2501
F-test (1st stage), (log) Property Taxes pp 0.93481
F-test (1st stage), p-value, (log) Property Taxes pp 0.33362
Wu-Hausman 4.3794
Wu-Hausman, p-value 0.03638
Wald (1st stage), (log) Property Taxes pp 0.73637
Wald (1st stage), p-value, (log) Property Taxes pp 0.39083
F-test (2nd stage) 5.4972
F-test (2nd stage), p-value 0.01905
Wald (2nd stage) 0.92402
Wald (2nd stage), p-value 0.33643
model 3
Dependent Var.: (log) Educ. Exp. pp
(log) Property Taxes pp 0.5876*
(0.2702)
Fixed-Effects: -------------------
fips Yes
year Yes
________________________________________ ___________________
S.E.: Clustered by: fips
Observations 51,490
R2 0.75440
Within R2 -0.26064
F-test (1st stage), (log) Property Taxes pp 3.5892
F-test (1st stage), p-value, (log) Property Taxes pp 0.02763
Wu-Hausman 2.9865
Wu-Hausman, p-value 0.08396
Wald (1st stage), (log) Property Taxes pp 2.8169
Wald (1st stage), p-value, (log) Property Taxes pp 0.05980
F-test (2nd stage) 6.7876
F-test (2nd stage), p-value 0.00918
Wald (2nd stage) 4.7306
Wald (2nd stage), p-value 0.02964
model 4
Dependent Var.: (log) Elementary Educ. Exp. pp
(log) Property Taxes pp 0.6125*
(0.2744)
Fixed-Effects: ------------------------------
fips Yes
year Yes
________________________________________ ______________________________
S.E.: Clustered by: fips
Observations 51,490
R2 0.74449
Within R2 -0.32920
F-test (1st stage), (log) Property Taxes pp 3.5892
F-test (1st stage), p-value, (log) Property Taxes pp 0.02763
Wu-Hausman 3.4873
Wu-Hausman, p-value 0.06185
Wald (1st stage), (log) Property Taxes pp 2.8169
Wald (1st stage), p-value, (log) Property Taxes pp 0.05980
F-test (2nd stage) 7.4623
F-test (2nd stage), p-value 0.00630
Wald (2nd stage) 4.9827
Wald (2nd stage), p-value 0.02561
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Changes in activity
# Using total active production as an instrument for per capita property tax
lapply(list(feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ diff_total_active_prod, data = mines_df, se = "cluster", cluster = "fips"),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ diff_total_active_prod, data = mines_df, se = "cluster", cluster = "fips"),
# Using the lag of total active production as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(diff_total_active_prod,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(diff_total_active_prod,1), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)), summary) %>% etable()NOTE: 2,710 observations removed because of NA values (IV: 0/2,710).
NOTE: 2,710 observations removed because of NA values (IV: 0/2,710). model 1
Dependent Var.: (log) Educ. Exp. pp
(log) Property Taxes pp 17.23
(167.9)
Fixed-Effects: -------------------
fips Yes
year Yes
________________________________________ ___________________
S.E.: Clustered by: fips
Observations 56,910
R2 -171.93
Within R2 -926.12
F-test (1st stage), (log) Property Taxes pp 0.00136
F-test (1st stage), p-value, (log) Property Taxes pp 0.97055
Wu-Hausman 1.3595
Wu-Hausman, p-value 0.24363
Wald (1st stage), (log) Property Taxes pp 0.01030
Wald (1st stage), p-value, (log) Property Taxes pp 0.91918
F-test (2nd stage) 1.2910
F-test (2nd stage), p-value 0.25587
Wald (2nd stage) 0.01052
Wald (2nd stage), p-value 0.91830
model 2
Dependent Var.: (log) Elementary Educ. Exp. pp
(log) Property Taxes pp 18.22
(177.4)
Fixed-Effects: ------------------------------
fips Yes
year Yes
________________________________________ ______________________________
S.E.: Clustered by: fips
Observations 56,910
R2 -192.52
Within R2 -1,050.9
F-test (1st stage), (log) Property Taxes pp 0.00136
F-test (1st stage), p-value, (log) Property Taxes pp 0.97055
Wu-Hausman 1.5382
Wu-Hausman, p-value 0.21489
Wald (1st stage), (log) Property Taxes pp 0.01030
Wald (1st stage), p-value, (log) Property Taxes pp 0.91918
F-test (2nd stage) 1.4626
F-test (2nd stage), p-value 0.22653
Wald (2nd stage) 0.01055
Wald (2nd stage), p-value 0.91821
model 3
Dependent Var.: (log) Educ. Exp. pp
(log) Property Taxes pp -1.651
(3.341)
Fixed-Effects: -------------------
fips Yes
year Yes
________________________________________ ___________________
S.E.: Clustered by: fips
Observations 54,200
R2 -1.1271
Within R2 -10.084
F-test (1st stage), (log) Property Taxes pp 0.05353
F-test (1st stage), p-value, (log) Property Taxes pp 0.81704
Wu-Hausman 0.58931
Wu-Hausman, p-value 0.44269
Wald (1st stage), (log) Property Taxes pp 0.27244
Wald (1st stage), p-value, (log) Property Taxes pp 0.60170
F-test (2nd stage) 0.43432
F-test (2nd stage), p-value 0.50988
Wald (2nd stage) 0.24421
Wald (2nd stage), p-value 0.62119
model 4
Dependent Var.: (log) Elementary Educ. Exp. pp
(log) Property Taxes pp -1.815
(3.685)
Fixed-Effects: ------------------------------
fips Yes
year Yes
________________________________________ ______________________________
S.E.: Clustered by: fips
Observations 54,200
R2 -1.4759
Within R2 -12.083
F-test (1st stage), (log) Property Taxes pp 0.05353
F-test (1st stage), p-value, (log) Property Taxes pp 0.81704
Wu-Hausman 0.70253
Wu-Hausman, p-value 0.40194
Wald (1st stage), (log) Property Taxes pp 0.27244
Wald (1st stage), p-value, (log) Property Taxes pp 0.60170
F-test (2nd stage) 0.53142
F-test (2nd stage), p-value 0.46602
Wald (2nd stage) 0.24258
Wald (2nd stage), p-value 0.62235
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Using total active mines as an instrument for per capita property tax
lapply(list(feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ diff_total_active_n, data = mines_df, se = "cluster", cluster = "fips"),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ diff_total_active_n, data = mines_df, se = "cluster", cluster = "fips"),
# Using lag of total active mines as an instrument for per capita property tax
feols(log_Total_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(diff_total_active_n,1:2), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year),
feols(log_Elem_Educ_Total_Exp_pp ~ 1 | fips + year | log_Property_Tax_pp ~ l(diff_total_active_n,1:2), data = mines_df, se = "cluster", cluster = "fips", panel.id = ~fips+year)), summary) %>% etable()NOTE: 5,420 observations removed because of NA values (IV: 0/5,420).NOTE: 5,420 observations removed because of NA values (IV: 0/5,420). model 1
Dependent Var.: (log) Educ. Exp. pp
(log) Property Taxes pp 0.2499
(0.3253)
Fixed-Effects: -------------------
fips Yes
year Yes
________________________________________ ___________________
S.E.: Clustered by: fips
Observations 56,910
R2 0.83289
Within R2 0.10408
F-test (1st stage), (log) Property Taxes pp 1.2430
F-test (1st stage), p-value, (log) Property Taxes pp 0.26490
Wu-Hausman 0.01524
Wu-Hausman, p-value 0.90175
Wald (1st stage), (log) Property Taxes pp 5.3428
Wald (1st stage), p-value, (log) Property Taxes pp 0.02081
F-test (2nd stage) 0.24782
F-test (2nd stage), p-value 0.61862
Wald (2nd stage) 0.59038
Wald (2nd stage), p-value 0.44228
model 2
Dependent Var.: (log) Elementary Educ. Exp. pp
(log) Property Taxes pp 0.1829
(0.3206)
Fixed-Effects: ------------------------------
fips Yes
year Yes
________________________________________ ______________________________
S.E.: Clustered by: fips
Observations 56,910
R2 0.83680
Within R2 0.11291
F-test (1st stage), (log) Property Taxes pp 1.2430
F-test (1st stage), p-value, (log) Property Taxes pp 0.26490
Wu-Hausman 7.21e-5
Wu-Hausman, p-value 0.99322
Wald (1st stage), (log) Property Taxes pp 5.3428
Wald (1st stage), p-value, (log) Property Taxes pp 0.02081
F-test (2nd stage) 0.13434
F-test (2nd stage), p-value 0.71398
Wald (2nd stage) 0.32540
Wald (2nd stage), p-value 0.56838
model 3
Dependent Var.: (log) Educ. Exp. pp
(log) Property Taxes pp 0.2703
(0.4858)
Fixed-Effects: -------------------
fips Yes
year Yes
________________________________________ ___________________
S.E.: Clustered by: fips
Observations 51,490
R2 0.82758
Within R2 0.11499
F-test (1st stage), (log) Property Taxes pp 0.61169
F-test (1st stage), p-value, (log) Property Taxes pp 0.54244
Wu-Hausman 0.01196
Wu-Hausman, p-value 0.91293
Wald (1st stage), (log) Property Taxes pp 1.9747
Wald (1st stage), p-value, (log) Property Taxes pp 0.13881
F-test (2nd stage) 0.24471
F-test (2nd stage), p-value 0.62083
Wald (2nd stage) 0.30943
Wald (2nd stage), p-value 0.57803
model 4
Dependent Var.: (log) Elementary Educ. Exp. pp
(log) Property Taxes pp 0.2461
(0.4838)
Fixed-Effects: ------------------------------
fips Yes
year Yes
________________________________________ ______________________________
S.E.: Clustered by: fips
Observations 51,490
R2 0.83037
Within R2 0.11758
F-test (1st stage), (log) Property Taxes pp 0.61169
F-test (1st stage), p-value, (log) Property Taxes pp 0.54244
Wu-Hausman 0.00497
Wu-Hausman, p-value 0.94381
Wald (1st stage), (log) Property Taxes pp 1.9747
Wald (1st stage), p-value, (log) Property Taxes pp 0.13881
F-test (2nd stage) 0.20536
F-test (2nd stage), p-value 0.65043
Wald (2nd stage) 0.25885
Wald (2nd stage), p-value 0.61091
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Bartik
Bartik Instrument
ss <- readRDS(here("data/temp/shift_shares_base_01_05_11.RDS"))
ss %>%
#filter(!(substr(fips, 1,2) %in% c("02", "15", "51", "09", "24"))) %>%
pull(fips) %>%
unique %>%
setdiff(unique(mines_df$fips), .)character(0)ss_df <- mines_df %>%
left_join(., ss, by = c("year", "fips"))
# library(ShiftShareSE)
# # Estimate the shift-share instrumental variable regression
# ivreg_ss(d_sh_empl ~ 1 | shock,
# X=IV,
# data=ADH$reg,
# W=ADH$W,
# method=c("ehw", "akm", "akm0")
# )
#
# ivreg_ss(log_Elem_Educ_Total_Exp_pp ~ 1 | fd_annual_avg_emplvl_21,
# X = fd_annual_avg_emplvl_21,
# data=ss_df,
# W=ss_df$share_emp_extraction,
# method=c("ehw", "akm", "akm0")
# )
ss_labs <- as_labeller(c("ss_emp_coal_2001" = "Coal",
"ss_emp_oil_gas_2001" = "Oil & Gas",
"ss_emp_extraction_2001" = "Extractive Industries",
"ss_wage_coal_2001" = "Coal",
"ss_wage_oil_gas_2001" = "Oil & Gas",
"ss_wage_extraction_2001" = "Extractive Industries"))
sswage <- ss_df %>%
select(fips, year,ss_emp_coal_2001, ss_emp_oil_gas_2001, ss_emp_extraction_2001) %>%
pivot_longer(contains("ss")) %>%
group_by(fips) %>%
filter(any(!is.na(value))) %>%
ungroup %>%
ggplot(aes(x = year, y = value, group = fips)) +
geom_line(alpha = 0.2, color = "darkblue") +
facet_wrap(~name, scales = "free", labeller = ss_labs) +
xlab("Year") +
ylab("Change in Total Employment") +
labs(title = "Employment-based Shift-Share Instrument") +
theme_bw() +
theme(legend.position = "none",
panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
strip.text = element_text(face = "bold"),
strip.background = element_rect(
color="grey", fill="lightsteelblue2", linetype="solid"),
panel.border = element_rect(color = "grey"),
plot.background = element_blank(),
plot.title = element_text(hjust=0.5))
ss_coal <- ss_df %>%
select(fips, year,ss_wage_coal_2001, ss_wage_oil_gas_2001, ss_wage_extraction_2001) %>%
pivot_longer(contains("ss")) %>%
group_by(fips) %>%
filter(any(!is.na(value))) %>%
ungroup %>%
mutate(value = value/1000000) %>%
ggplot(aes(x = year, y = value, group = fips)) +
geom_line(alpha = 0.2, color = "violetred3") +
facet_wrap(~name, scales = "free", labeller = ss_labs) +
xlab("Year") +
ylab("Change in Total Wages (millions)") +
labs(title = "Wage-based Shift-Share Instrument") +
theme_bw() +
theme(legend.position = "none",
strip.background = element_rect(
color="grey", fill="pink1", linetype="solid"),
panel.grid.major = element_blank(),
panel.border = element_rect(color = "grey"),
strip.text = element_text(face = "bold"),
panel.grid.minor = element_blank(),
plot.background = element_blank(),
plot.title = element_text(hjust=0.5))
bartik <- arrangeGrob(grid.arrange(sswage, ss_coal, ncol = 1))Warning: Removed 39459 rows containing missing values (`geom_line()`).
Removed 39459 rows containing missing values (`geom_line()`).
ggsave(here("output/plots/bartik_graphs.jpg"), bartik, width = 25, height = 25, units = "cm")Wage Instrument
scaled_ss <- ss_df %>%
mutate(across(contains("ss"), ~./1000000000))
feols(log_Elem_Educ_Total_Exp_pp ~ log_gdp_total_pc +
sw(ss_wage_coal_2001, ss_wage_coal_2005, ss_wage_coal_2011,
ss_wage_oil_gas_2001, ss_wage_oil_gas_2005, ss_wage_oil_gas_2011,
ss_wage_ff_2001, ss_wage_ff_2005, ss_wage_ff_2011,
ss_wage_extraction_2001, ss_wage_extraction_2005, ss_wage_extraction_2011) | fips + year, data = scaled_ss, se = "cluster", cluster = "fips")Standard-errors: Clustered (fips)
Expl. vars.: log_gdp_total_pc + ss_wage_coal_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.230071 0.071927 3.19868 0.0026264 **
ss_wage_coal_2001 -0.088016 0.075890 -1.15979 0.2526849
---
Expl. vars.: log_gdp_total_pc + ss_wage_coal_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.209336 0.070373 2.974676 0.0043219 **
ss_wage_coal_2005 -0.057379 0.072852 -0.787609 0.4342466
---
Expl. vars.: log_gdp_total_pc + ss_wage_coal_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.147750 0.061732 2.393406 0.020337 *
ss_wage_coal_2011 -0.000336 0.064127 -0.005233 0.995845
---
Expl. vars.: log_gdp_total_pc + ss_wage_oil_gas_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.217069 0.033905 6.402266 8.0818e-10 ***
ss_wage_oil_gas_2001 0.005454 0.038958 0.140005 8.8877e-01
---
Expl. vars.: log_gdp_total_pc + ss_wage_oil_gas_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.219827 0.028286 7.771541 1.2246e-13 ***
ss_wage_oil_gas_2005 -0.017032 0.032431 -0.525183 5.9984e-01
---
Expl. vars.: log_gdp_total_pc + ss_wage_oil_gas_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.222985 0.031451 7.089853 8.4996e-12 ***
ss_wage_oil_gas_2011 -0.021485 0.053896 -0.398640 6.9042e-01
---
Expl. vars.: log_gdp_total_pc + ss_wage_ff_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.505203 0.252155 2.003543 0.10148
ss_wage_ff_2001 -0.065494 0.073039 -0.896705 0.41097
---
Expl. vars.: log_gdp_total_pc + ss_wage_ff_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.374696 0.093686 3.999489 0.0011605 **
ss_wage_ff_2005 -0.012152 0.029666 -0.409635 0.6878637
---
Expl. vars.: log_gdp_total_pc + ss_wage_ff_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.395129 0.114553 3.449305 0.0035769 **
ss_wage_ff_2011 -0.001196 0.029609 -0.040405 0.9683035
---
Expl. vars.: log_gdp_total_pc + ss_wage_extraction_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.239140 0.021478 11.13401 < 2.2e-16 ***
ss_wage_extraction_2001 -0.004966 0.003418 -1.45296 0.14653
---
Expl. vars.: log_gdp_total_pc + ss_wage_extraction_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.240558 0.020146 11.94065 < 2.2e-16 ***
ss_wage_extraction_2005 -0.009496 0.003782 -2.51081 0.012178 *
---
Expl. vars.: log_gdp_total_pc + ss_wage_extraction_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.245755 0.018330 13.40741 < 2.2e-16 ***
ss_wage_extraction_2011 -0.014311 0.003647 -3.92361 9.2067e-05 ***Employment Instrument
feols(log_Elem_Educ_Total_Exp_pp ~ log_gdp_total_pc +
sw(ss_emp_coal_2001, ss_emp_coal_2005, ss_emp_coal_2011,
ss_emp_oil_gas_2001, ss_emp_oil_gas_2005, ss_emp_oil_gas_2011,
ss_emp_ff_2001, ss_emp_ff_2005, ss_emp_ff_2011,
ss_emp_extraction_2001, ss_emp_extraction_2005, ss_emp_extraction_2011) | fips + year, data = scaled_ss, se = "cluster", cluster = "fips")Notes from the estimations:
[x 1] The variable 'ss_emp_ff_2001' has been removed because of collinearity (see $collin.var).Standard-errors: Clustered (fips)
Expl. vars.: log_gdp_total_pc + ss_emp_coal_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.232853 0.072495 3.21198 0.002531 **
ss_emp_coal_2001 -17323.823187 11399.084124 -1.51976 0.136065
---
Expl. vars.: log_gdp_total_pc + ss_emp_coal_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.210717 0.070703 2.980326 0.0042538 **
ss_emp_coal_2005 -10867.409061 11381.054732 -0.954868 0.3437482
---
Expl. vars.: log_gdp_total_pc + ss_emp_coal_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.154212 0.062471 2.468539 0.016891 *
ss_emp_coal_2011 -4441.761829 9865.117918 -0.450249 0.654402
---
Expl. vars.: log_gdp_total_pc + ss_emp_oil_gas_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.215446 0.03374 6.385460 8.8768e-10 ***
ss_emp_oil_gas_2001 11472.594487 16874.75758 0.679867 4.9725e-01
---
Expl. vars.: log_gdp_total_pc + ss_emp_oil_gas_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.217589 0.028187 7.719600 1.7222e-13 ***
ss_emp_oil_gas_2005 3201.326352 13930.618763 0.229805 8.1840e-01
---
Expl. vars.: log_gdp_total_pc + ss_emp_oil_gas_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.222334 0.031361 7.089606 8.5128e-12 ***
ss_emp_oil_gas_2011 -4157.499655 20845.055784 -0.199448 8.4204e-01
---
Expl. vars.: log_gdp_total_pc + ss_emp_ff_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.36961 0.090687 4.07567 0.0095805 **
---
Expl. vars.: log_gdp_total_pc + ss_emp_ff_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.381064 0.102464 3.719006 0.0020574 **
ss_emp_ff_2005 -4877.147951 10010.817440 -0.487188 0.6331681
---
Expl. vars.: log_gdp_total_pc + ss_emp_ff_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.421735 0.133923 3.149089 0.0066187 **
ss_emp_ff_2011 -4385.346879 11202.275518 -0.391469 0.7009553
---
Expl. vars.: log_gdp_total_pc + ss_emp_extraction_2001
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.238894 0.021392 11.16731 < 2.2e-16 ***
ss_emp_extraction_2001 -898.369720 707.180882 -1.27035 0.20424
---
Expl. vars.: log_gdp_total_pc + ss_emp_extraction_2005
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.239842 0.02007 11.95029 < 2.2e-16 ***
ss_emp_extraction_2005 -1705.995400 853.73025 -1.99828 0.045915 *
---
Expl. vars.: log_gdp_total_pc + ss_emp_extraction_2011
Estimate Std. Error t value Pr(>|t|)
log_gdp_total_pc 0.245383 0.018344 13.37701 < 2.2e-16 ***
ss_emp_extraction_2011 -2619.261079 788.660033 -3.32115 0.00092282 ***Summary statistics with coal production
# Variables of interest for descriptive statistics
vars <- c("Total_Revenue", "Property_Tax", "Total_IG_Revenue", "Total_Fed_IG_Revenue", "Total_State_IG_Revenue", "Total_Expenditure")
mines_df %>%
mutate(across(c(esa_tot_exp, vars, paste0(vars, "_pp"), gdp_total, gdp_priv_ind, gdp_govt, gdp_o_g_mining_quarr_21), ~./1000)) %>%
select(year, esa_tot_exp, esa_tot_exp_pp, Enrollment, all_of(vars), paste0(vars, "_pp"), pop_total, pop_school_age, gdp_total, gdp_priv_ind, gdp_govt, gdp_o_g_mining_quarr_21, total_active_n, total_active_prod) %>%
# rename("ESA Expenditure" = esa_tot_exp,
# "ESA Expenditure per pupil" = esa_tot_exp_pp,
# "Revenue" = "Total_Revenue",
# "Property Tax" = "Property_Tax",
# "IG Revenue" = "Total_IG_Revenue",
# "Federal IG Revenue" = "Total_Fed_IG_Revenue",
# "State IG Revenue" = "Total_State_IG_Revenue",
# "Expenditure" = "Total_Expenditure",
# "Revenue per pupil" = "Total_Revenue_pp",
# "Property Tax per pupil" = "Property_Tax_pp",
# "IG Revenue per pupil" = "Total_IG_Revenue_pp",
# "Federal IG Revenue per pupil" = "Total_Fed_IG_Revenue_pp",
# "State IG Revenue per pupil" = "Total_State_IG_Revenue_pp",
# "Expenditure per pupil" = "Total_Expenditure_pp",
# "Population" = pop_total,
# "School Age Population" = pop_school_age,
# "GDP" = gdp_total,
# "Private Industry GDP" = gdp_priv_ind,
# "Government GDP" = gdp_govt,
# "Oil, gas, mining GDP" = gdp_o_g_mining_quarr_21) %>%
#dplyr::group_split(year < 2007) %>%
# purrr::iwalk(.,~{
# .x %>%
select(-year) %>%
data.frame() %>%
stargazer(digits = 0, digits.extra = 3, covariate.labels = c("ESA Expenditure",
"ESA Expenditure per pupil",
"Enrollment",
"Revenue",
"Property Tax",
"IG Revenue",
"Federal IG Revenue",
"State IG Revenue",
"Expenditure",
"Revenue per pupil",
"Property Tax per pupil",
"IG Revenue per pupil",
"Federal IG Revenue per pupil",
"State IG Revenue per pupil",
"Expenditure per pupil",
"Population",
"School Age Population",
"GDP",
"GDP - Private Industry",
"GDP - Government",
"GDP - Oil, gas, mining",
"Active Coal Mines",
"Coal Produced (k short tons)"), type = "latex")
% Table created by stargazer v.5.2.3 by Marek Hlavac, Social Policy Institute. E-mail: marek.hlavac at gmail.com
% Date and time: Thu, Mar 21, 2024 - 11:27:35
\begin{table}[!htbp] \centering
\caption{}
\label{}
\begin{tabular}{@{\extracolsep{5pt}}lccccc}
\\[-1.8ex]\hline
\hline \\[-1.8ex]
Statistic & \multicolumn{1}{c}{N} & \multicolumn{1}{c}{Mean} & \multicolumn{1}{c}{St. Dev.} & \multicolumn{1}{c}{Min} & \multicolumn{1}{c}{Max} \\
\hline \\[-1.8ex]
ESA Expenditure & 7,648 & 20 & 37 & 0 & 463 \\
ESA Expenditure per pupil & 7,648 & 1 & 2 & 0 & 37 \\
Enrollment & 56,910 & 14,654 & 50,465 & 2 & 1,723,961 \\
Revenue & 56,910 & 169 & 614 & 0 & 25,052 \\
Property Tax & 56,910 & 63 & 252 & 0 & 7,909 \\
IG Revenue & 56,910 & 94 & 370 & 0 & 18,363 \\
Federal IG Revenue & 56,910 & 1 & 4 & 0 & 137 \\
State IG Revenue & 56,910 & 91 & 361 & 0 & 17,810 \\
Expenditure & 56,910 & 172 & 628 & 0 & 23,008 \\
Revenue per pupil & 56,910 & 0 & 0 & 0 & 0 \\
Property Tax per pupil & 56,910 & 4 & 5 & 0 & 165 \\
IG Revenue per pupil & 56,910 & 7 & 3 & 0 & 116 \\
Federal IG Revenue per pupil & 56,910 & 0 & 0 & 0 & 0 \\
State IG Revenue per pupil & 56,910 & 0 & 0 & 0 & 0 \\
Expenditure per pupil & 56,910 & 0 & 0 & 0 & 0 \\
Population & 56,910 & 95,455 & 324,316 & 256 & 10,105,708 \\
School Age Population & 56,910 & 19,360 & 66,325 & 45 & 2,219,357 \\
GDP & 56,910 & 4,758 & 20,754 & 4 & 836,163 \\
GDP - Private Industry & 56,910 & 4,180 & 18,726 & 3 & 752,486 \\
GDP - Government & 56,910 & 579 & 2,186 & 1 & 83,677 \\
GDP - Oil, gas, mining & 56,910 & 95 & 572 & 0 & 27,694 \\
Active Coal Mines & 56,910 & 0 & 2 & 0 & 92 \\
Coal Produced (k short tons) & 56,910 & 340 & 6,575 & 0 & 415,924 \\
\hline \\[-1.8ex]
\end{tabular}
\end{table}